Performance Evaluation of Branch and Value Prediction Using Discrete-Event Simulation of Fluid Stochastic Petri Nets

Mitrevski, Pece and Gušev, Marjan (2001) Performance Evaluation of Branch and Value Prediction Using Discrete-Event Simulation of Fluid Stochastic Petri Nets. In: Proceedings of the Second Conference on Informatics and Information Technology. Institute of Informatics, Faculty of Natural Sciences and Mathematics, Ss. Cyril and Methodius University in Skopje, Macedonia, Skopje, Macedonia, pp. 39-56. ISBN 9989-668-28-0

[img]
Preview
Text
9989-668-28-0_pp39-56.pdf

Download (309kB) | Preview
Official URL: http://ciit.finki.ukim.mk

Abstract

Speculative execution is one of the key issues to boost the performance of future generation microprocessors. In this paper, Fluid Stochastic Petri Nets are used to capture the dynamic behavior of an ILP processor, and discrete-event simulation is applied to assess the efficiency of predictions and speculative execution in boosting the performance of ILP processors that fetch, issue, execute and commit a large number of instructions per cycle. The evaluation leads to numerous conclusions regarding the performance impact of branch and value prediction under different scenarios, with varying parameters of both the microarchitecture and the operational environment – branch and value prediction accuracy, machine width, instruction window size and operational profile.

Item Type: Book Section
Uncontrolled Keywords: Instruction Level Parallelism, speculative execution, Branch Prediction, Value Prediction, Fluid Stochastic Petri Nets, discrete-event simulation
Subjects: International Conference on Informatics and Information Technologies > Computer architectures
Depositing User: Vangel Ajanovski
Date Deposited: 28 Oct 2016 00:15
Last Modified: 28 Oct 2016 00:15
URI: http://eprints.finki.ukim.mk/id/eprint/11252

Actions (login required)

View Item View Item